------------------------------------------------------------------------------- ************************************** NIRffs, version 1.0 (14-February-2009) ************************************** INDEX 1 - PROGRAM DESCRIPTION 1.1 - Introduction 1.2 - Error estimation 2 - INSTALLATION 3 - PROGRAM USE 3.1 - Description 3.2 - Examples 4 - REFERENCES ------------------------------------------------------------------------------- 1 - PROGRAM DESCRIPTION ======================= 1.1 - Introduction ------------------ This program predicts the strength of the Ca triplet indices (CaT*, PaT and CaT; defined in Cenarro et al. 2001) and the MgI and sTiO indices (Cenarro et al. 2009) on the basis of the empirical fitting functions computed in Cenarro et al. (2002) and Cenarro et al. (2009). It performs the required interpolations between the different local functions as explained in the above two papers. Therefore, it is a very simple program specially designed to be easily implemented into stellar populations synthesis codes. This software is copyright by J. Gorgas, A. J. Cenarro and N. Cardiel. 1.2 - Error estimation ---------------------- Together with the indices predictions, the program also computes the random errors associated to such predictions resulting from the covariance matrices of the fits (for the indices CaT*, PaT, MgI and sTiO). This ensures a reliable error index estimation for any combination of input atmospheric parameters. See more details on errors estimation in Cenarro et al. (2002; 2009) 2 - INSTALLATION ================ You only need the adequate FORTRAN compiler for your OS. Install the program by executing: % g77 -o output_name NIRffs.f (Linux) % f77 -o output_name NIRffs.f (Alpha, Sparc) NOTE: Although we provide a complet program on its own, obviously, the main program can be deleted and the subroutine "predicted()" (and their associated subroutines) can be included in any separate code. 3 - PROGRAM USE =============== 3.1 - Description ----------------- When you execute the program, you get the following message: "Input Teff, log(g), [Fe/H]" You just have to type the three atmospheric parameters - effective temperature, in K - logarithm (base 10) of the surface gravity, in dex - metallicity, in dex all of them in the same line and between comas. It returns the INDICES CaT*, PaT, CaT, MgI and sTiO (first column; all of them in angstroms except for sTiO, which is adimensional), the corresponding ERRORS (second column; in the corresponding index units) and an integer CODE indicating the success and reliability of the prediction (third column). Notes: - Output CODES: 0 -> OK. Safe prediction. 1 -> Extrapolation to high/low Teff and/or log(g) -1 -> Error (Teff out of valid range, see below; null indices returned) -2 -> Error ([Fe/H] out of valid range, see below; null indices returned) -3 -> Error (log(g) out of valid range, see below; null indices returned) -5 -> Error - We have established the following valid parameter ranges: 2240 K < Teff < 100800 K (0.05 < theta < 2.25) -1.0 < logg < 6.0 -4.5 < [Fe/H] < 1.5. For input values out of the above regimes, the corresponding output codes will be returned. Note however that, for each particular index, there are certain parameter ranges in which the fitting functions do not depend on gravity and/or metallicity (see Cenarro et al. 2002, 2009), and an index value can still be computed even if the input parameter is out of the valid regime (see examples in Section 3.2). 3.2 - Examples -------------- (a) INPUT --> 4550.,2.30,-0.70 OUTPUT --> CaT* = 6.677 0.132 0 PaT = 0.638 0.036 0 CaT = 7.270 0.137 sTiO = 0.956 0.002 0 MgI = 0.558 0.029 0 In this case, the codes "0" are indicating that the predicted indices are OK. (b) INPUT --> 5000.,5.20,0.00 OUTPUT --> CaT* = 6.368 0.439 1 PaT = 0.594 0.095 1 CaT = 6.920 0.448 sTiO = 0.912 0.011 1 MgI = 1.012 0.051 1 The codes "1" indicate that, for this region of the parameter space, the gravity is out of the library coverage. Note that this could not be the case for other parameters regions as in the next example (c). (c) INPUT --> 3500.,5.20,0.00 OUTPUT --> CaT* = 4.826 0.131 0 PaT = 0.277 0.101 0 CaT = 5.084 0.161 sTiO = 1.005 0.010 0 MgI = 0.692 0.044 0 (d) INPUT --> 3800.,1.,999. OUTPUT --> CaT* = 10.159 0.772 0 PaT = 0.735 0.151 0 CaT = 10.843 0.785 sTiO = 0.000 0.000 -2 MgI = 0.000 0.000 -2 In this case (d), the input [Fe/H] is out of the valid range. The codes '0' for CaT* and PaT indicate that the prediction is still OK, as their fitting function predictions do not depend on [Fe/H] at 3800 K. On the other hand, since MgI and sTiO predictions do depend on [Fe/H] at this Teff, a null index value and a code '-2' is returned. As it is illustrated in the next examples (e) and (f), this can change depending on the input Teff. (e) INPUT --> 3500.,1.,999. OUTPUT --> CaT* = 8.292 0.662 0 PaT = 1.495 0.187 0 CaT = 9.682 0.684 sTiO = 1.250 0.027 0 MgI = 0.000 0.000 -2 (f) INPUT --> 3100.,1.,999. OUTPUT --> CaT* = 3.494 0.333 1 PaT = 1.803 0.231 1 CaT = 5.170 0.396 sTiO = 2.514 0.043 1 MgI = 0.373 0.075 1 In case (e), at 3500 K, sTiO fitting functions do not depend on [Fe/H], but MgI ones still do. Finally (f), at 3100 K no fitting function depend on [Fe/H] so the four indices can be predicted (are non-null). For such a low Teff, log(g) is somewhat larger (out of the library coverage), hence the code '1' appears. 4 - REFERENCES ============== - Cenarro A.J., Cardiel N., Gorgas J., Peletier R.F., Vazdekis A., Prada F., 2001, MNRAS, 326, 959 - Cenarro A.J., Gorgas J., Cardiel N., Vazdekis A., Peletier R.F., 2002, MNRAS, 329, 863 - Cenarro A.J., Cardiel N., Vazdekis A., Gorgas J., 2009, MNRAS, submitted